{"id":"W3150202690","doi":"10.48550/arxiv.2103.16617","title":"HAD-Net: A Hierarchical Adversarial Knowledge Distillation Network for Improved Enhanced Tumour Segmentation Without Post-Contrast Images","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Segmentation; Computer science; Artificial intelligence; Context (archaeology); Modality (human–computer interaction); Inference; Contrast (vision); Dice; Machine learning; Pattern recognition (psychology); Mathematics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.000200708,0.0004571582,0.0004981628,0.0001376526,0.0004178471,0.0002310534,0.00116799,0.0002937678,0.00001124946],"category_scores_gemma":[0.00008033566,0.0005485383,0.0003157941,0.000692606,0.0001373023,0.0006349977,0.00114682,0.0005832938,0.00001455279],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003378398,"about_ca_system_score_gemma":0.0003641967,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003514567,"about_ca_topic_score_gemma":0.0001848984,"domain_scores_codex":[0.9970836,0.0001964863,0.0003849761,0.001657274,0.00008998086,0.0005876541],"domain_scores_gemma":[0.9973838,0.0003937375,0.0004782412,0.0009858946,0.0005332337,0.0002250945],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006225734,0.0003835353,0.0005335059,0.0001896954,0.0003111736,0.00003390085,0.0007576778,0.8675055,0.04390573,0.06496725,0.0006878368,0.02010163],"study_design_scores_gemma":[0.001833457,0.0001360229,0.002468545,0.0001206547,0.0001598998,0.000004683889,0.00007365632,0.9637861,0.004168405,0.02613003,0.0003202709,0.0007982315],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06499869,0.00009430692,0.9313059,0.000265751,0.0009927263,0.001588863,0.00008243382,0.0003294366,0.0003418768],"genre_scores_gemma":[0.9360479,0.00009653532,0.06202252,0.00009860132,0.0006546768,0.0000386464,0.0004608233,0.00004073768,0.0005396079],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8710492,"threshold_uncertainty_score":0.9996966,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03022948599163722,"score_gpt":0.2225361786316715,"score_spread":0.1923066926400342,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}